LFM3-Omni-2B Runs Fully Offline on iPhone at 80 tok/s

The xenovacom team ported LiquidAI's multimodal LFM3-Omni-2B to run fully offline on iPhone 18 Pro Max at roughly 80 tok/s, using custom WebGPU kernels in a React Native app for local full-duplex multimodal chat without separate ASR/TTS components.

2026-11-05 ~ 2026-11-05 · 3 related posts

• LFM3-Omni-2B runs fully offline on iPhone at ~80 tok/s via custom WebGPU kernels — xenovatech · 2026-11-05
• LFM3-Omni-2B ported to run fully offline on iPhone at ~80 tok/s via custom WebGPU kernels — victormustar · 2026-11-05
• LFM3-Omni-2B ported to run fully offline on iPhone at ~80 tok/s with custom WebGPU kernels — JosephJacks_ · 2026-11-05